Title: Applying metaheuristic approaches on the single facility location problem with polygonal barriers

Authors: Hossein Gharehmeshk Gharravi; Mohammad Saleh Farham

Addresses: Department of Mathematics and Statistics, Curtin University, Perth, 6102, Australia ' Department of Industrial Engineering, Middle East Technical University, Ankara, 06800, Turkey

Abstract: In this paper, we describe, improve, and compare the performance of four metaheuristic algorithms on the single facility location problem with barriers. The barriers are polygonal regions on a plane where neither facility location nor travelling is permitted. The objective function in this problem is non-convex justifying the application of metaheuristics and their robustness. We argue that this problem has interesting geometric properties such as size and distribution of barriers on the plane which can affect the choice of metaheuristics. Various test instances which reflect these geometric properties are designed in this paper. Through computational experiments, we discuss the effect of these properties on different variants of metaheuristics and propose the suitable metaheuristic and parameter setting for the single facility location problem with barriers.

Keywords: polygonal barriers; facility location; metaheuristics; swarm intelligence; Weber problem; simulated annealing; artificial bee colony; ABC; gravitational search algorithm; GSA; elite-based algorithm.

DOI: 10.1504/IJMHEUR.2014.068915

International Journal of Metaheuristics, 2014 Vol.3 No.4, pp.348 - 370

Accepted: 11 Feb 2015
Published online: 25 Apr 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article